COVID-19: Countries Wealth, Cases, Deaths and Vaccines

Motivation \ Coronavirus is one of the common viruses that can cause infection in your sinuses, nose or upper throat. Most of them are not dangerous as they present with mild symptoms and can be treated symptomatically, but it killed around 858 people from Middle East respiratory syndrome (MERS) in 2015, this is because of its severe presentation causing respiratory failure. Undiscovered coronaviruses like the Coronavirus are very dangerous because specific treatment for such viruses is not yet available and it rapidly progresses to cause multiorgan failure. This category of the virus causes harmful diseases in mammals as well as in birds. In humans, the virus causes mild respiratory infections, which in rare cases may even cause death. In animals like cows and pigs, it causes diarrhea, while in chickens it causes severe respiratory infections. You may be shocked to know that there are no vaccines currently available for the treatment of this disease.During the Covid-19 outbreak, lots of amazing dashboards were released. However, I was particularly curious about how the wealth of a country affects the main indicators like the number of cases, deaths and vaccines. Hope this post helps people gain some insight on that as well!

Exploratory Data Analaysis on

  1. Cases, Deaths, Vaccinations.
  2. Reproduction Rate.
  3. Stringency Index.
  4. Vaccines used.

Questions

  1. Do Wealthy countries have less cases?
  2. Do Wealthy countries have less deaths?
  3. Do Wealthy countries have more vaccines? Did they have them earlier?

Data Source

Special thanks to Our World in Data for making available such a complete, clean, updated and reliable data as they are doing with such an important topic.And the last date in the dataset is Yesterday(2022/5/13)\ Our World in Data (https://ourworldindata.org/ - https://github.com/owid/covid-19-data/tree/master/public/data)

Importing Libraries

Datasets

We know that we have a total of 220 countries.\ The first date in the dataset is January 1st of 2020.\ The last date in the dataset is Yesterday\ The key is Country-Date, meaning that there is only 1 row for each day and country.\ More info about the columns of the dataset can be found in: \ https://github.com/owid/covid-19-data/blob/master/public/data/owid-covid-codebook.csv \ https://github.com/owid/covid-19-data/tree/master/public/data

I will now split the df in two dataframes, one country related and another one country&date. Meaning that one will be used for data particularly related with covid evolution and the other one with parameters from the country itself.

Country indicators

From this plot we can see that we have some missing values in the population_density, gdp_per_capita, life_expectancy and human_development_index but no so many. Meaning that this dataframe, the variables per countries are quite complete.

Country-Date cases, deaths, tests, vaccinations, reproduction rate, stringency index

Thanks to this plot, we can easily se that there are a lot of missing values for ICU/hospitalized patience. This may be explained due to several countries not reporting it. That is why those columns will not be used further in this notebooks and will be dropped straight away. \ Similarly, we can see high number of missing values in the columns related with the vaccinations. In this case, the reason is not the same as before, instead this could be explained due to vaccines only available in 2021, but covid started since 2019/20. \ Furthermore, we can also see missing values in the tests columns. This can be explained due to two factors, some countries did not report the tests performed at all, and some other reported in a less frequent base than daily (for example, weekly).

This plots are quite simple, but provides a quick overview of the current situation of Covid. Where we can see that vaccines, finally, vaccinations have overcame - by far - the quantity of cases. \ Parallelly, the pie chart on the right hand, show small porcentage (yet relevant and important) of deaths among positive cases reported.

Vaccinations

This plot tell us the countries that have applied the most quantity of vaccines to date. It is biased due to the population of each of them. The bigger the country, the more vaccinations, but not necessarily is the one that is doing that best.

Similar to previous plot, but this time it is normalized by the population of the country. So one can really see which countries are doing it best (possibly smaller countries find it easier/faster to vaccinate the whole population).

Now we can see the same information as in the previous plot, but in a map which makes it easy have all the information in one view.

From this plot we can see which countries are currently (from last 7 days) vaccinating at a higher pace, considering the population of the country.

From this map we can see which countries are vaccinating more, considering the lasts 7 days and the population of the countries.

Scatters

From this plot we can see the relation between the total vaccinations and the total vaccinations considering the population of the country. We can't identify a relation with the continents. It's mainly descriptive, but not much to infer from this plot.

Since the colors represent the continents, we can identify certain aligment withing the European country, meaning that countries in Europe seems to be vaccinating at a relatively similar pace when it comes to daily vaccinations, as well as the accumulated vaccinations. \ We can also identify that African countries seem to be in the bottom left of the chart, which is unfortunately low vaccinations accumulated and daily (except Seychelles). \ Apart from this two observations, the rest of the datapoints seem dispersed to draw conclusions only with it.

With this treemap, we can see how much a vaccination schema is used in total. This treemap is a bit misleading due to few countries reporting how many dosis per vaccine were applied (per day or total), hence if a country is applying 10% of a certain vaccine and 90% of another one (supposing only 2 for this example), by looking at the plot you may thing that it is applying 50% of each when that's not the case.

This plot is similar to the one before, but considering the population of the countries

Cases

Here we can have an overall view of the covid cases per million habitant. We can see highly affected areas in South America, Europe and USA.

This plot pretends to show which are the countries where currently present higher daily cases.

Evolution over Time of tests, cases, deaths and vaccinations

With this plot we can see the overall figures evolution across time.

GDP, Treemaps

From this coloured treemap, I was expecting to see some relation with the GDP, accumulated cases and the continents but it is not the case. There is not trend I can highlight.

This treemap, shows which countries are currently having more daily cases, grouped by continents. It is similar to the "Daily new cases per million habitants" map, but in here we can see better the scales. In the map, small countries could hardly be seen in a color scale, while here its size is easier to understsand and call attention.

Stringency Index

This is a composite measure based on nine response indicators including school closures, workplace closures, and travel bans, rescaled to a value from 0 to 100 (100 = strictest). If policies vary at the subnational level, the index is shown as the response level of the strictest sub-region.

Above animated map, shows how the severity of restrictions evolved in each of the countries.

Reproduction Rate

Above animated map, shows how the reproduction rate evolved in each of the countries.

Correlation

This heatmap plots the Pearson correlation indexes for the different main indicators. Again, I was epecting to see more cases or deaths in less develop / low-income countries, but that is not the case (or not strong enought to my eyes).

Questions

  1. Do Wealthy countries have less cases?
  2. Do Wealthy countries have less deaths?
  3. Do Wealthy countries have more vaccines? Did they have them earlier?

1. Do Wealthy countries have less cases?

From the above two plots, we can see that the wealthier the country, the earlier it reported Covid19 cases. We can also see that around May2021, there is a drastic reduction on the high-income countries, possible due to the vaccines. From the second chart, we can see that the high-income countries show more quantity of population tested positive.

2. Do Wealthy countries have less deaths?

When it comes to deaths, we can see a big and early spike for high-income countries with a drastically reduction around July2020. Again, in January2021 they presented more deaths than other lower income countries. Something that caught my attention in the two plots, is the significantly lower deaths in the lowest income countries. A possible explanation could be that the figures do not reflect what actually happened / is happening in reality.

3. Do Wealthy countries have more vaccines? Did they have them earlier?

For vaccinations, we can see again higher numbers for high-income countries. It is not only clear that they are having more vaccines than the lower groups, but also that they started to have them earlier. This may be attributable, among other causes, to the fact that some high-income countries were the firsts ones to develop the vaccine (prioritizing their own population), or also that these countries were able to pay higher prices in order to get the shots sooner.

Conclusion

To sum up, wealthier countries reported a higher number of cases, deaths and vaccines. Furthermore, the wealthier the country, the earlier the vaccination processes started.

Disclaimer

Even though the Our World in Data team made a fantastic job gathering and wrangling the data, the conclusions could not reflect the reality due to countries measuring/reporting in different ways, applying different tests criteria’s, etc.